Welcome to the CATS!

navigation icon Info

navigation icon Outline

navigation icon Homework

navigation icon MATLAB Examples

navigation icon Exams and Projects

navigation icon Course Home

nystar logo

Rensselaer logo

ECSE 6480 Adaptive Systems (Spring 2011)

Text: I will draw on the following books that are available for download from the web:

* Karl J. Astrom and Bjorn Wittenmark, Adaptive Control , 2nd Ed., Addison Wesley, 1995. (main source for the course.) Downloadable from Prof. Wittenmark's webpage.
* Marc Bodson and Shankar Sastry, Adaptive Control: Stability, Convergence, and Robustness, Prentice-Hall, 1994. Downloadable from Prof. Bodson's webpage
* Petros Ioannou and Jing Sun, Robust Adaptive Control, 1996. Downloadable from Prof. Ioannou's webpage.

Prerequisites: ECSE 6400 (Systems Analysis Techniques) or equivalent.

Course website (RPI LMS): https://rpilms.rpi.edu/webct/logon/165361262001

Course Objective: The course covers the theory and application of adaptive control. The course considers mostly linear time invariant systems, and also some specific classes of nonlinear systems. The topics covered in the course include: what adaptation is and when it is needed, off-line and real-time parameter estimation algorithms, direct and indirect adaptive methods, deterministic self-tuning regulators, Lyapunov stability theory, input-output stability, model-reference adaptive control, adaptive observer, stability and convergence of adaptive algorithms, and robustness issues. Adaptive algorithms will be developed in both continuous time domain and discrete time domain. Gain scheduling, implementation aspects of adaptive control, and applications to physical systems will also be discussed.

Learning Outcomes: Students who complete this course satisfactorily will be able to: (i) apply off-line and on-line parameter estimation algorithms to linear time invariant systems and specific classes of nonlinear systems (ii) apply feedback adaptive observer and control for the regulation and tracking for dynamical systems (iii) analyze robustness of adaption algorithms.

Grade Composition:

Final Exam

Homework: Homework will account for 15% of the course grade.  Problems will be assigned roughly each week. The problems are best done individually in a professional manner (neatness counts!).  Problems will be generally be graded on a 10 point scale and will be returned in class approximately one week after they were due.  Collaboration in the solution of the homework problems is permitted and is strongly encouraged if it enhances the learning process, but mere copying of the solution is not allowed.  In general, no late assignments will be accepted.  However, extensions may be granted if a situation arises for which it is warranted.  In these instances the student must request the extension in writing prior to the assignment due date, stating the reason for the request and the date the assignment is to be submitted.

Project: Project will account for 20% of the course grade. Up to two persons may collaborate on a project, but both must sign a statement describing the respective contribution. The project report is due on 5/18 by noon in CII 8011.

Homework and project may require the use of MATLAB and associated toolboxes.  

Exam: Both midterm and final exams will be take-home exams. The midterm will be handed out in class on 3/21 and due at 5pm on 3/25 (in Instructor's office, CII 8011). The final exam will be handed out on 5/9 (last class) and due on 5/16 by noon in CII 8011. Take home exams may involve the use of computation tools (e.g., MATLAB). ABSOLUTELY no collaboration allowed for the take-home exams. Collaboration on the take-home exams constitutes cheating.

Statement of Academic Integrity
Student-teacher relationships are built on trust. The students must trust that the instructor has made appropriate decisions about the structure, content, etc., of the courses they teach, and the instructors must trust that assignments which students turn in are their own. Acts which violate this trust undermine the education process.
The Rensselaer Handbook defines various forms of Academic Dishonesty and procedures for dealing with them. All forms are violations of the trust between the students and instructors. All students should familiarize themselves with this portion of the Rensselaer Handbook and should note that the penalties for the various forms of dishonesty can be quite harsh. Cheating on the take-home exams will result in a zero score, and referral to the Dean of Students for possible additional action.

Additional References
Goodwin and Sin, Adaptive filtering prediction and control, Prentice-Hall, 1984.
Miroslav Krstic Ioannis Kanellakopoulos Petar Kokotovic, Nonlinear and Adaptive Control Design, John Wiley and Sons, 1995.
P.A. Ioannou and B. Fidan, Adaptive Control Tutorial, SIAM – Society for Industrial & Applied Mathematics, TJ217.I628, ISBN 0-89871-615-2, 2006.

Course Coverage (Total number of classes: 28)
Parametric models
Parameter identification in dynamical models
Adaptive pole placement
Model reference adaptive control
Adaptive cotnrol of nonlinear systems
Gain scheduling

Back to top